YCB-Ev

YCB-Ev数据集由弗劳恩霍夫IGD创建，包含21个同步的事件和RGB-D序列，总时长7:43分钟，用于评估6DoF物体姿态估计算法。该数据集提供21个YCB物体的真实6DoF姿态数据，与YCB-Video数据集中的物体相同，支持跨数据集算法性能评估。数据集创建过程中，使用Intel RealSense D435和Prophesee EVK2相机进行同步捕捉，确保数据质量和同步性。YCB-Ev数据集主要应用于增强现实、虚拟现实和机器人领域的实时物体检测和6D姿态估计，旨在解决现有算法在真实世界条件下的泛化能力问题。

YCB-Ev dataset was developed by Fraunhofer IGD. It contains 21 synchronized event and RGB-D sequences with a total duration of 7 minutes and 43 seconds, and is designed for evaluating 6DoF object pose estimation algorithms. This dataset provides ground-truth 6DoF pose data for 21 YCB objects, which are identical to those included in the YCB-Video dataset, supporting cross-dataset performance evaluation of algorithms. During its construction, synchronized data capture was conducted using Intel RealSense D435 and Prophesee EVK2 cameras to ensure data quality and synchronization. The YCB-Ev dataset is primarily applied to real-time object detection and 6D pose estimation in the fields of augmented reality, virtual reality and robotics, aiming to address the generalization ability issue of existing algorithms under real-world conditions.